Complexity: When Computer Models Simply Will Not Do

Biological systems are too complex to be well-modeled by computers, currently. Medical personnel are on the front lines, saving lives. Society requires a far higher level of certainty of results from medical research than can be achieved by today's computer models.

Most biological models make use of lab animals such as mice or rats. But increasingly, acellular models are used. A fascinating type of biochip that allows placing various cellular components on a silicon dioxide surface--connecting them by microfluidic channels--promises to provide increasingly ingenious acellular models for biomedical research.

Now, in a major breakthrough, a group of researchers at the Weizmann Institute of Science in Israel, led by Roy Bar-Ziv, in collaboration with Margherita Morpurgo from the University of Padova in Italy, have designed a molecule affectionately called the “daisy” that is able to bind genes onto chips in miniature patterned arrays.

Bar-Ziv and co-workers have been able to use the daisy to pattern tiny regions of double-stranded DNA onto silicon dioxide surfaces. Indeed, these immobilized genes are able to conduct their business on patterned silicon substrates without the need for living cells. These biochips can act as protein microtraps, selectively trapping specific proteins from crude cell extracts with high spatial resolution. Moreover, the gene sequences immobilized on the biochips can be used for the on-chip production of proteins by transcription/translation processes such as those occurring within cells.

Bar-Ziv and his colleagues have also demonstrated the integration of these systems with microfluidics. Integration with flow systems is of interest for the fabrication of miniature assembly lines on chips, wherein proteins can be synthesized on the chips and transported to their final destinations through microfluidic channels.

In a remarkable demonstration of the utility of the daisy approach, the researchers have patterned two different genes as alternating stripes on a biochip. The protein synthesized on one stripe diffuses to the second stripe where it regulates the synthesis of a second protein. More complex artificial gene circuits can be envisioned by extending this protocol, and thus the biochips may be able to carry out complex cascaded information-processing functions, mimicking those in living organisms.

Source.


While computer modeling may be sufficient for political organisations such as the IPCC and Greenpeace to predict an apocalyptic future for the earth, for down-to-earth bio-researchers who absolutely must provide reliable results, computers are simply not good enough. But bio-engineers, bio-physicists, and molecular biologists working with nanotechnologists are growing increasingly clever in designing useful models that do not use lab animals.

It will be many years before biomedical research no longer needs to sacrifice large numbers of lab animals for the saving of human life. But it is encouraging to see the creation of "artificial cells" on silicon, and it is fascinating to contemplate the possibilities they offer.